On stabilizing reinforcement learning without Lyapunov functions.
Pavel OsinenkoGrigory YaremenkoGeorgiy MalaniyaPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- nonlinear systems
- adaptive control
- stability analysis
- reinforcement learning algorithms
- optimal policy
- dynamical systems
- model free
- temporal difference
- learning algorithm
- state space
- least squares
- supervised learning
- closed loop
- function approximation
- optimal control
- control law
- learning process
- multi agent